Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case

Testo completo

(1)ROBERT SCHUMAN CENTRE FOR ADVANCED STUDIES. EUI Working Papers RSCAS 2010/21 ROBERT SCHUMAN CENTRE FOR ADVANCED STUDIES European Report on Development. FRAGILE STATES, COMMODITY BOOMS AND EXPORT PERFORMANCE: AN ANALYSIS OF THE SUB-SAHARAN AFRICAN CASE. Andrew Mold and Annalisa Prizzon.

(2)

(3) EUROPEAN UNIVERSITY INSTITUTE, FLORENCE ROBERT SCHUMAN CENTRE FOR ADVANCED STUDIES. EUROPEAN REPORT ON DEVELOPMENT. Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case ANDREW MOLD AND ANNALISA PRIZZON. EUI Working Paper RSCAS 2010/21.

(4) This text may be downloaded only for personal research purposes. Additional reproduction for other purposes, whether in hard copies or electronically, requires the consent of the author(s), editor(s). If cited or quoted, reference should be made to the full name of the author(s), editor(s), the title, the working paper, or other series, the year and the publisher.. ISSN 1028-3625. © 2010 Andrew Mold and Annalisa Prizzon Printed in Italy, April 2010 European University Institute Badia Fiesolana I – 50014 San Domenico di Fiesole (FI) Italy www.eui.eu/RSCAS/Publications/ www.eui.eu cadmus.eui.eu.

(5) Robert Schuman Centre for Advanced Studies The Robert Schuman Centre for Advanced Studies (RSCAS), directed by Stefano Bartolini since September 2006, is home to a large post-doctoral programme. Created in 1992, it aims to develop inter-disciplinary and comparative research and to promote work on the major issues facing the process of integration and European society. The Centre hosts major research programmes and projects, and a range of working groups and ad hoc initiatives. The research agenda is organised around a set of core themes and is continuously evolving, reflecting the changing agenda of European integration and the expanding membership of the European Union. Details of this and the other research of the Centre can be found on: http://www.eui.eu/RSCAS/Research/ Research publications take the form of Working Papers, Policy Papers, Distinguished Lectures and books. Most of these are also available on the RSCAS website: http://www.eui.eu/RSCAS/Publications/ The EUI and the RSCAS are not responsible for the opinion expressed by the author(s).. European Report on Development The European Report on Development (ERD) is a multiannual process, which is framed within the initiative “Mobilizing European Research for Development Policies“. The objective of this initiative is to enhance a European perspective on development issues in the international arena, on the basis of knowledge excellence, innovation and building of common ground between the European research community and policy-makers (Member States and Commission). Moreover, this initiative will improve the visibility of the EU at a global level, help shape the international agenda and feed the EU internal debate on development. The ERD Team Leader is Prof. Giorgia Giovannetti. The multidisciplinary team is based at the Robert Schuman Centre of the European University Institute and is interacting with a broad network of scholars, from both developed and developing countries. http://erd.eui.eu/ For further information: European Report on Development Robert Schuman Centre for Advanced Studies European University Institute Via delle Fontanelle, 19 50016 San Domenico di Fiesole (FI), Italy Fax: + 39 055 4685 770 E-mail: erdsec@eui.eu.

(6)

(7) Abstract Sub-Saharan Africa’s export performance over recent decades has typically been por-trayed as poor compared to other regions in developing countries. This paper takes a new look at the record, using data on the volume rather than the value of African ex-ports. When analysed in volume terms a different picture of African export performance emerges. Despite being confronted by sharply declining prices, between 1995-2001 Afri-can exports expanded by an average of 5.9 percent annually. The picture changes quite significantly during the post-2002 commodity price boom period, with increases of 5.2 per cent per annum in average volumes. By using a dynamic panel of 36 Sub-Saharan countries, the aim of this paper is to analyse this apparent paradox, using data available from UNCTAD. Specifically, we investigate the price-elasticity response of African ex-porters in the light of dramatically shifting unit prices. In the context of the EDR project, we also specifically look at the question of whether countries classified as ´fragile states´ have been especially disadvantaged in terms of their export performance.. Keywords Export determinants, Sub-Saharan Africa, export performance, commodity prices.

(8)

(9) 1. Introduction*1 In recent decades, it has not been difficult to find statements lamenting Africa’s purportedly poor export performance. For example, in 2005, the Commission for Africa Report suggested that: “[…]The last three decades have seen stagnation in Africa. The composition of Africa’s exports has essentially remained unchanged, and has contributed to a collapse in Africa’s share of world trade.[…] Africa will not be able to achieve the Millennium Development Goals, nor set itself on a sustainable path to growth and poverty reduction, without increased trade.” (Commission for Africa, 2005:256).. Is this kind of affirmation justified? This paper takes a new look at the record, using data on the volume, rather than the value, of Sub-Saharan African exports, and in the light of what has been, up until a few months ago, the most sustained commodity price boom since the early 1900s (World Bank, 2008) (Figure 1). Figure 1: Real Prices for Commodities over the Long-Run, 1900-2008, yr2000=100. Up until the early 2000s, one of the most important stylised facts of African exports is that exporters had been facing a sharp decline in the price of most of their commodities. World prices for many of the commodities that Africa exports declined sharply between 1990 and 2000: cocoa, cotton, sugar and copper by over 25 per cent, coffee by 9 per cent and minerals overall declined by 14 per cent (WTO, 2001: 212). Product price instability was also a major problem for exporters. One half of traditional products experienced average price changes of 50 per cent or more during the 1990s (Ng and Yeats, 2002). Price changes were associated with the collapse of traditional product prices, and this reduced the value of export earnings.. *. 1. Paper prepared for the Workshop on “Food crisis and the development potential of the agricultural sector in fragile countries”, organised by the European Report of Develop-ment in Cambridge, United Kingdom, 17-18 March 2009. The views expressed herein are solely those of the authors and do not necessarily represent the views of the OECD. They contributed this paper in their personal capacities. The authors would like to thank Flavine Creppy (UNCTAD) for collaboration with providing data, Augustin Fosu and Oliver Morrissey for the comments on an earlier draft of this paper, and also all the participants at the Cambridge ERD Workshop held in March 2009. Any errors are of course the responsibility of the authors.. 1.

(10) Andrew Mold and Annalisa Prizzon. Yet, despite being confronted by sharply declining prices, between 1990-2001, the volume of exports for Sub-Saharan African non-oil exporters actually increased by an average of 5 per cent annually. This impressive supply-side performance has not been properly documented: previous studies having fixed too much attention to the value of African exports, something which, as primary commodity exporters, is largely beyond their control. Subsequent developments were largely unexpected. Commodity prices surged from 2001-8, in the longest sustained boom since the early 1900s (World Bank, 2008). Indeed, while, for Africa as a whole, export unit prices fell by 2 per cent per annum between 1995 and 2001, they increased at a yearly rate of 17 per cent between 2002 and 2006 (UNCTAD, 2008a). Many analysts forecast a longterm super-cycle for commodity prices, and the structural long-term problems of low prices and high volatility of commodity prices were largely forgotten. How did African country exporters respond to this new international context and heavy demand for many of its traditional exports? Ironically enough, export performance for African countries (as measured in volumes) actually improves less rapidly with increases of 5.2 per cent in average volumes between 2002-2006, despite surging prices. Put bluntly, the supply-side response in the face of the commodity price boom was disappointing, especially when contrasted with the fairly strong performance that was evident in the period from 1980-2001. The aim of this paper is to analyse the performance of African exports at the aggregate level in volume terms, using data available from UNCTAD. Specifically, we investigate the price-elasticity response of African exporters in the light of dramatically shifting unit prices. In the context of the EDR project, we also specifically look at the question of whether countries classified as “fragile states” have been especially disadvantaged in terms of their export performance. The structure of the paper is as follows. Section 2 presents some of the stylised facts on African exports. Section 3 provides a literature review of the supply side response of African exporters, both in the agricultural and non-agricultural sectors. Section 4 discusses the policy environment in which African trade performance has to be considered. Section 5 presents the main econometric results. Section 6 concludes by drawing some policy implications.. 2. Africa’s Trade Performance in Perspective – The Key Role of Prices One of the most extensively cited stylised facts of African trade performance is that the continent’s share in world merchandise trade, measured in value terms, has declined steadily since 1980, from around 6 per cent to around 2 per cent in the late 1990s, with a subsequent mild recovery (due mainly to the recovery in prices of some key exports) in the 2000s to around 3 per cent. This decline in the world share of exports has been particularly marked for Western and Southern Africa. However, as Morrissey (2005: 1134) argues, this does not mean that trade is unimportant for Africa: compared to other developing country regions, Sub-Saharan Africa (SSA) tends to have high export/GDP and import/GDP ratios. In simple terms, exports are very important to African countries even if African exports are not very important in the world market. Economists are divided on the significance of Africa’s declining share in world markets (Gibbon and Ponte, 2005: 38). Some present deeply pessimistic assessments – Sachs and Warner (1997), for example, portray Africa’s trade performance as having been an unmitigated disaster. Others, however, present a more nuanced interpretation. Using a gravity model specification, Rodrik (1999) arrives at the conclusion that, when other variables affecting trade are controlled for (such as location and per capita income), “there is little evidence that trade policies have repressed trade volumes below cross-national benchmarks, unless they have done so indirectly through their depressing effect on incomes…..once the focus is shifted from trade to economic growth in general, we are forced to think more broadly about the whole range of growth determinants, and not just the impediments to exchanges at the border.” (ibid., page 113).. 2.

(11) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. For example, from 1995 to 2001, a period still characterised by low commodity prices, many African countries still managed to expand their export volumes significantly (Table 1), increasing at an average annual rate of 5.9 per cent per annum, only slightly less than for Africa than for the developing world as a whole (7.6 per cent). In contrast, during the period of high commodity prices since 2002, export performance was actually worse if looked at from the point of view of volumes, expanding at an annual average rate of 5.2 per cent. In particular, despite the boom in commodity prices, export volume performance in Eastern and Middle Africa2 markedly worsens: annual growth rate of export volume amounts to 11.3 per cent and 9 per cent for Middle African countries in the periods 1995-2001 and 2002-2006 respectively; 7.7 per cent and -0.3 in the case of Eastern African countries. The key difference between the two periods from 1995-2001 and 2002-2006 is that, while export prices decreased for African countries by -2.1 per cent per annum in the first period, they increased by 19.1 per cent per annum in the second period.3 The story is a similar one in the case of Latin America where, despite an annual growth rate of 7.7 per cent in the export price index over the period 2002-2006, the export volume index increased by only 5.7 per cent. In contrast, over the period 1995-2001, the export volume annually grew by 9.4 per cent while the annual growth rate of the export price index was negative (-2.2 per cent). Taking into account sample composition (see Annex A.3.), the case of Oceania is emblematic: despite an upsurge in the export unit value index of 20.1 per cent in the period 2002-2006, the growth rate of the export volume index is actually negative (-0.6). Prices, then, play a key role in explaining the difference in export performance between the two periods. Table 1: Value, Volume, Price Indices, 1995-2006 (% Annual Change) 1995-2001. 2002-2006. Value indices of exports. Volume indices of exports. Unit value indices of exports. World. 3.0. 5.9. -2.7. Developing economies. 5.0. 7.6. -2.4. Developing economies: Africa. 3.6. 5.9. -2.1. 25.3. 5.2. 19.1. Eastern Africa. 1.1. 7.7. -6.1. 14.3. -0.3. 14.7. Middle Africa. 5.6. 11.3. -5.1. 35.8. 9.0. 24.5. Northern Africa. 5.6. 5.2. 0.4. 28.1. 6.1. 20.7. Southern Africa. 0.8. 4.1. -3.2. 18.4. 5.6. 12.1. Western Africa. 3.3. 3.0. 0.4. 24.5. 4.0. 19.7. Developing economies: America. 7.3. 9.4. -2.0. 18.1. 5.7. 11.7. Developing economies: Asia. 4.7. 7.4. -2.6. 22.0. 13.3. 7.7. 5.4. 20.2. -0.6. 21.0. Developing economies: Oceania -4.2 -9.1 Source: Calculated from UNCTAD Handbook of Statistics, 2008. Volume indices of exports. Unit value indices of exports. 16.9. 7.7. 8.5. 21.6. 11.2. 9.4. Value indices of exports. It is often suggested that the key export problem for developing countries has not been the prices of commodities per se, but rather their volatility. It is certainly the case that some countries have been very negatively affected by price volatility – it can play havoc with government revenues and adversely impact on public expenditure in areas such as health, infrastructure and education. It is also often argued that with the trend towards the globalisation of markets and increasing liberalisation,. 2 3. See Annex A.3. for country composition of regional aggregates. These price increases in the 2002-6 period might seem quite moderate in view of the strong rise in prices in 2007 and the first half of 2008. When the financial markets weakened from mid-2007, investors moved much of their money into the markets for oil, gold, cereals and other commodities. However, the UNCTAD data is not yet available for these two periods.. 3.

(12) Andrew Mold and Annalisa Prizzon. volatility has increased (for example, Rodrik, 1999). In terms of the key commodities for African producers, this is not necessarily correct - on average price volatility declined significantly over the course of the 1970s, 1980s and 1990s for all commodities except for tropical beverages, though it did increase again quite significantly during the boom of the 2000s (Figure 2). Figure 2: Commodity Price Instability4 – 1970-2007 (annual average percentage variation). Source: Calculated from UNCTAD Handbook of Statistics, 2008. The problems of volatility, important as they are, should not detract from the underlying problems of a secular decline in the terms of trade. Ever since the controversial and path-breaking studies of Singer (1950) and Prebisch (1950), a debate has raged over the net barter terms of trade for commodity exporters vis-à-vis the exporters of manufactured goods. Some authors have tried to dismiss the hypothesis, pointing out (correctly) that the trend depends on the period chosen, and on the relative importance of primary commodities in the total composition of developing country exports.5 Nevertheless, over the long run, the basic hypothesis has been repeatedly tested and found valid (for example, Spraos, 1983; Bloch and Sapsford, 2000, Ocampo and Parra, 2003). During the commodity boom of recent years, such debates have largely been overlooked, as developing country exporters have enjoyed a period of strong prices. Serious concerns over the global “food crisis”, as food prices soared over recent years, have also contributed to overshadowing the issue of long-term price trends. Yet it would be wrong to take the recent commodity boom out of perspective. For hard commodities, many of the increases have been the largest seen since the commodity boom of the 1970s. But for “soft commodities”, upon which many low-income countries rely, this has not been the case (for example, Figure 3).6. 4 5. 6. The price instability index is calculated as the average percentage deviation of prices from their exponential trend. For instance, in his otherwise exhaustive text book Feenstra (2004:338) dispatches the hypothesis with the simple oneline affirmation that “there is little evidence to support that hypothesis in general”. It is worth noting that the increasing terms of trade in favour of African countries between 2003 and 2008 was large, by 7.8 per cent, mainly due to gains by oil exporting countries, which amounts to 16.9 per cent. But the gains by non-oil exporters were very low, at just a meagre 1.1 per cent (IMF, 2008).. 4.

(13) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. Figure 3: Agricultural Raw Material Prices 1960-2008. Source: Calculated from UNCTAD Handbook of Statistics, 2008. Notes: Constant prices have been calculated by deflating by the Unit Price for Manufactured Goods.. Although the issue has become complicated in the past decade by a sharply declining trend in manufacturing products (on which we will comment more in a moment), theoretical analysis suggests several reasons why agricultural commodity prices fall relative to others. Firstly, because of relatively inelastic demand and the lack of differentiation among producers, which means that agricultural markets are highly competitive. On the supply side, technological improvements, increased competition, reduced protection of markets and devaluation of some national currencies (for example, CFA franc) of many agricultural commodity-producing countries (following structural adjustment programmes) have all contributed to significant increases in production. On the demand side, the development of synthetic substitutes further displace agricultural commodities as intermediate inputs, reducing further the growth in demand. In the past, these trends have undermined Africa’s trade performance in value terms. For particular countries, these broad trends had a particularly perverse impact. Wuyts (2005), for example, documents that from 1987 to 2001 Tanzania’s overall terms of trade (for goods and services) dropped by nearly 30 per cent, while the terms of trade for goods dropped by nearly 40 per cent. The agricultural terms of trade declined by even more, about 50 per cent. Wuyts (2005:11) concludes that: “poverty incidence may rise, even if per capita GDP is growing (without adverse changes in income inequality) because declining external terms of trade may offset the gains from increased production.”. None of this is to deny that other developing regions have also had to contend with declining terms of trade. This is, of course, to some extent the inevitable (and desirable) consequence of rising productivity (which, moreover, tends to be higher in the tradeable sector, rather than in non-tradeable, which are less exposed to international competition) which feeds in through in lower unit costs. Thus, countries such as China, which have had an enormous success in expanding labour intensive manufacturing exports have had to face sharply declining export prices, too. An UNCTAD study showed, for instance, that, despite the country’s considerable success in promoting exports, China’s net barter terms of trade (that is, the prices of their exports compared to their imports) in manufactures deteriorated by more than 10 per cent over the period 1993-2000 (Zheng and Yumin, 2002). These countries have been particularly vulnerable to the problems associated with the “fallacy of composition” – the idea that, if a number of developing countries simultaneously try to increase exports in a similar range of product categories, then they may all end up losing through insufficient foreign demand and depressed international prices. However, some countries, such as South Korea, have successfully managed to avoid this dilemma by increasingly shifting production towards higher value-added, technologically more sophisticated, dynamic manufactures. A study by Berge and. 5.

(14) Andrew Mold and Annalisa Prizzon. Crowe (1997, cited in UNCTAD, 2003:90), for instance, reveals no significant trends in the net barter terms of trade of the Republic of Korea regarding its trade in manufactures with advanced industrial countries, but a significant increase vis-à-vis other developing countries and an even greater increase in the income terms of trade, suggesting a relatively successful strategy of diversification into higher value-added products, compared to its less developed trading partners. The tragedy of Africa is, of course, that countries in the continent have not managed a similar transformation. We will discuss briefly why this has not been the case in Section 4.. 3. Literature Review on the African Supply Side Response to Price Shifts The literature assessing the price elasticity of production and export – especially in the agricultural sector - mainly focuses on country studies. But doubts have been raised about the reliability of such studies (for example, Mamingi, 1996, Ogbu and Getibuou, 1990). Ogbu and Gebtibouo (1990) argue that the empirical literature on agricultural supply behaviour in SSA is characterised by models that are deficient either in their methodology or in their choice of relevant explanatory variables. This is because the models fail to recognise the structure of agricultural production in these countries and over-simplify the issues and constraints facing farmers. They also criticise the assumption that the binding constraints are uniform across countries. According to Chhibber (1988), cross-country estimations of supply functions suffer from the problem of establishing the direction of causality: the assumption underlying supply functions is that prices influence output or productivity when, in fact, it is possible to argue that high agricultural productivity, which is associated with high per capita incomes, may lead to higher price support for agriculture. Ogbu and Gebtibouo (1990) argue that a negative price elasticity is often associated with the target income hypothesis and backward bending supply curve – in other words, producers respond to declines in price by producing more in an attempt to sustain their income levels. They suggest that factors that influence productivity such as technology, fertilizer use, institutional support, access to credit, health services and infrastructure are more likely to influence output supply responsiveness than price. In a review of agricultural supply response studies, Ozamme (1999:64) notes that: “Empirical results which do not have the ‘correct sign’ tend to be rejected and therefore go unreported in academic publications. The weight of empirical evidence may therefore be misleading and economists and policy-makers alike should be wary of accepting prevailing dogma unreservedly.”. Contributions to the literature on the relationship between prices and volumes, focusing specifically on aggregate export supply-side functions, are listed in Annex Table A.8. To summarise some of the findings, in the case of South Africa, Naudé (2000) finds that none of the determinants identified in the literature are significant in explaining South Africa’s export volume. McKay et al. (1998) and Cerra and Saxena (2002) base their analyses on price and volume indices. In the case of Tanzania, McKay et al. (1998) suggest that agricultural supply response is quite strong, so that the potential for agricultural sector response to liberalisation of agricultural prices and marketing may be quite significant. The long-run elasticity of food crop output to relative prices is almost unity. Both food and aggregate short-run response are estimated at about 0.35. Cerra and Saxena (2002) study the determinants of China’s export volume, with mixed results regarding the influence of price on volumes. To sum up, broadly speaking, despite the misgivings of authors such as Ozanne (1999), previous micro-level studies on agricultural supply side response in SSA tend to show a small positive effect of prices on volume.. 4. The Role of Trade Policy in Explaining African Export Performance An important stylised fact, at odds with conventional wisdom, is that African countries are surprisingly open to international trade. Measured by the standard index of “openness” of African economies (i.e., exports plus imports as a percentage of GDP), African countries typically display an openness ratio in the order of 50-60 per cent, comparable to the average of the European Union. 6.

(15) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. countries, and about three times higher than that displayed by the world’s biggest importer, the United States. This implies that, all African countries could now be considered as “open”. And as a recent World Bank study notes: “tariffs have been falling throughout much of the region over the past decade for capital goods, intermediate goods and total imports. […]Furthermore, tariff rates do not appear any higher in Africa than in the more successful exporters in East and South Asia” (Clarke, 2005:7).7. For SSA countries on average, scheduled tariffs have fallen from 33 per cent in the early 1980s to 15 per cent by 2002 (Morrissey, 2005: 1139). The majority of SSA countries have implemented significant liberalisation of trade since the 1980s, with reforms mainly related to reducing restrictions on imports (Ackah and Morrissey, 2005). In theory, trade liberalisation reduces relative price distortions, encouraging an expansion of exports, which promotes economic growth. Although the empirical evidence linking trade liberalisation to growth is quite weak (Santos-Paulino, 2002a, 2002b), in SSA, there does appear to have been an export response (relative to GDP): imports have risen fairly slowly and, on average, export growth has tended to match this (Ackah and Morrissey, 2005). There is some evidence that growth has been higher in the more open SSA economies (Onafowora and Owoye, 1998), even those dependent on primary commodity exports (Mbabazi, Milner and Morrissey, 2003). However, as observed above, in SSA, export values tend to be determined externally, by trends in world commodity prices, and export earnings are variable across countries and over time, depending on the commodities that they export (Morrissey, 2005). Countries dependent on primary (agricultural) exports have a limited ability to increase production in response to improved price incentives following trade liberalisation (Mckay, Morrissey and Vaillant, 1997; Noorbakhsh and Paloni, 1998). Furthermore, SSA countries have failed to expand exports of manufactures, largely due to inefficiency and a lack of investment in technology in African manufacturing firms (Söderbom and Teal, 2003). The existence of many fragile states across the continent has compounded all these difficulties. Consequently, export response to trade liberalisation has been at best slow. Arguably, some of the reforms instigated compounded, rather than alleviated, some of the aforementioned difficulties associated with the declining terms of trade for African exports. In the area of agricultural policy, for instance, during the 1980s and 1990s, much pressure was placed upon African countries to abolish their marketing boards, on the grounds that they distorted price incentives in the agricultural sector, and had been used to “plunder agriculture”. By 1992, 16 marketing boards covering cash crops in 23 countries had given up their monopoly positions or had been eliminated (UNCTAD, 1999:ix). Although sometimes corrupt or inefficient, these boards provided information and facilities such as credit and extension services to farmers, and mobilised the country’s market power in selling the crop for export. Their elimination has regrettably left farmers vulnerable to the full force of price shocks in commodity markets. An empirical study by Boratav (2001), surveying evidence on 20 Sub-Saharan countries finds that de-regulation has not been associated with improvements in real producer prices or in the terms of trade. In an authoritative account of trends in the cocoa market, ul Haque (2004) documents the decline in cocoa prices subsequent to liberalisation. Together these stylised facts suggest that the basic problem is not that Africa trades too little, nor that the trade regime is excessively closed – rather, that it is trading the wrong kind of products: primary commodities with low valued-added, low prices and a very low elasticity of demand on. 7. Contrary to this view, some authors argue that the trade regimes of Africa are still extremely restrictive. Using an index of trade restrictiveness, Khandelwal (2004) suggests that only about 9 countries in the Eastern and Southern African region have open trade regimes, and these do not include the relatively developed countries in the region like Kenya or South Africa. Arguably, such a view confuses measurement of the policy environment with the actual outcome – although it is true that significant barriers to trade remain in most African countries, and these are relatively high by world standards, the simple fact that trade represents such a high percentage of GDP suggests that in practice these economies are indeed extremely open. The structural trade deficits which these countries sustain also corroborate the relatively open nature of the trade regimes.. 7.

(16) Andrew Mold and Annalisa Prizzon. world markets. Expressed alternatively, African producers have been unable to command “rents” from the production of their exports (Kaplinsky, 2005). The question of overcoming Africa’s economic marginalisation is, therefore, essentially qualitative, rather than quantitative. Seen from this perspective, the low (and decreasing) share of Africa in world trade is merely the outcome of a poor relative economic performance, and is reflected equally in other indicators such as Africa’s declining share of world GDP.8 In consonance with the analysis of Rodrik (1999) cited earlier, this conclusion is supported by econometric evidence from Coe and Hoffmaister (1998), who estimate a gravity model to address the question of whether Africa’s bilateral trade with the industrial countries is “unusual” compared with other developing country regions. Their main finding is that the unusually low level of African trade is fully explained by economic size, geographical distance, and population; if anything, the average African country tends to “overtrade” compared with developing countries in other regions. What emerges from this brief review of African trade policies is that countries have been following appropriate trade policies (i.e., export taxes and the most severe quantitative restrictions have been eliminated and tariffs have been reduced) but have not derived a significant benefit. Many reasons can and have been offered to explain this, especially natural barriers, high trade costs, structural characteristics and institutional weaknesses (for example, Morrissey, 2005). Overdependence on a (narrow range of) primary commodity exports has been recognised as a major constraint to export-led growth in SSA, and it is this issue that we now address in more detail.. 5. Econometric Analysis The export volume function can be described as follows (Fosu, 2003):. where VI corresponds to export volume, P to relative price, C the relative production cost and U is the perturbation element which includes systematic and random terms. We can identify foreign (f) and domestic (d) determinants:. The full unbalanced panel includes data for 36 Sub-Saharan African countries over the period 19802006 (covering the period of trade reforms and the subsequent commodity price boom). For our own analysis, we estimate the following model, each variable expressed in logs where feasible: where VI is specifically the volume index of exports and P is the unit price index for exports taken from the UNCTAD Handbook of Statistics 2008.9 The other control variables of our model were chosen to reflect better world demand, DW, domestic export distortions, TD, and supply-side capacity, Z, of the African economies under analysis. First, world demand is proxied by imports from High-income OECD Countries (World Development Indicators), which is expected to affect export volumes of African countries positively. Second, domestic export distortions are measured by REER, the real effective exchange rate, which assumes value 100 in the base year 2002 (International Finance Statistics). A depreciation of the REER (decrease in the real interest rate) is supposed to influence export competitiveness positively. Supply-side capacity is expressed by the following variables: GDP per capita in Local Currency Unit is used as a proxy for average productivity rates of labour (World Development Indicators) and FDI stock as a percentage of GDP (UNCTAD Handbook of Statistics 2008) tries to capture the growing involvement of multinationals in export-commodity. 8. 9. For instance, with over 11 per cent of the world’s population, Sub-Saharan Africa has barely one per cent of world’s GDP (UNCTAD Handbook of Statistics, 2005). This pattern is repeated in areas like FDI, were Africa’s share of total does not reach more than around 1 per cent of total inflows. See Mold (2004). Both indices are based on 2000 reference year, i.e. for the year 2000 the index equals 100.. 8.

(17) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. chains and the exploitation of natural resources. By using a lead dependent variable, all explanatory variables are effectively lagged with the exception of the proxy for average labour productivity. In order to investigate whether fragile countries present different behaviour in terms of export response, we consider an interaction term between the unit price index and a dummy variable that takes value 1 in the case of a fragile country. Specifically, a country is defined as fragile once it meets at least one of the fragility criteria presented in the ERD Outline Report.10 is an interaction factor between the unit price index and a time dummy which takes value one in the period 2002-2006 of corresponds to the sum of the country-specific term and the high commodity prices. Finally, disturbance term. Table 2: GMM Estimation on Export Volumes Dependent variable Volume Index Unit Price Index. Exports Volume Index (1). Exports Volume Index (2). 0.83*** (237.95) 0.23*** (42.43). 0.71*** (28.24) 0.12 (1.14) 0.05** (2.24) -0.11*** (-3.23) 0.56*** (5.22) 0.07*** (2.98) -0.30*** (-2.91) 0.01 (1.41). OECD Income REER GDP per capita (+1) FDI Stock Fragility interaction factor Time interaction factor. -0.25*** (-44.33) 0.02*** (12.19). Number of observation 1076 724 J-stat 42.12 26.35 S.E. of regression 0.3046 0.3070 Notes: t-statistics are reported under parenthesis. The levels of significance are as follows: *** 99 per cent level, **95 per cent level, *90 per cent level.. Given potential endogeneity problems, we estimate the reduced form dynamic panel using a GMM estimator (Arellano and Bond, 1991; Arellano and Bover, 1995). In the Annex we include the fixedeffects estimation for robustness checks, being aware of the biased co-efficient of the lagged dependent variable implied by this technique. Table 2 provides the main results. 5.1 Discussion of the Results Results of estimation (2) in Table 2 are suggestive, showing volume export performance to be highly ambiguous with respect to price: positive, but not significant. This conclusion is re-inforced by the fact that nor is the time interaction factor significant, suggesting no response to the strong prices of the post 2002 period. All the other variables are significant and of the expected signs. Average productivity, as proxied by GDP per capita, makes the largest impact on volumes. FDI stock is a significant determinant, in line with the view that FDI in SSA has traditionally been attracted to export activities (usually in the extractive sector) as the economies are mainly too small for market-. 10. The list of fragile countries is included in Annex A.4. The criteria are the following: LICUS (low-income country under stress), countries with a stateness score in Bertelsmann Index lower than 7, Failed State Index above 90, countries that fell into the bottom two CPIA quintiles (2003-2005-2006) and proxy list of fragile states provided by DFDI (2005).. 9.

(18) Andrew Mold and Annalisa Prizzon. seeking FDI. World demand (proxied by OECD imports) also plays a small, but positive, significant role in explaining export volumes. Consistent with the idea that overvalued exchange rates detract from export performance, REER exerts a significant negative impact on export volumes. Most importantly from the point of view of the ERD project, the fragility interaction factor is significant and negative, reflecting the special difficulties that fragile states encounter in terms of their ability to sustain their export performance. These results lead to an obvious question about why the response from rising commodity prices has been so ambiguous and is not picked up by our interaction dummy. A number of explanations are plausible: 1. As mentioned earlier, price rises have been concentrated disproportionately on certain products. It means little for countries whose major export items have been little affected by these trends (for example, tea in Kenya) or has experienced price movements for its own reasons, quite independently of any other markets (such as vanilla in Madagascar) (Lines, 2008:62). 2. Another possibility is that African exporters themselves have not been benefiting fully from the price rises. The argument has been made forcefully by Gibbon and Ponte (2005), and Kaplinsky (2005) among others: growing market concentration among purchasers means that markets behave more like monopsonies than competitive markets, and increases in market prices are not fully passed on to producers. 3. Another explanation is the way in which prices have been calculated. As commodity prices are typically denominated in dollars, their price increases are smaller in the currencies that appreciate against that currency. If the price increase is smaller in the currency of a commodity importing country, the demand response will also be smaller than in the absence of a dollar depreciation (UNCTAD Trade and Development Report, 2008b:22). That has, indeed, been the case in the last two-three years, as the US real exchange rate has declined.. 6. Conclusions The objective of this study has been to analyse the fundamental determinants of African trade performance. It has been argued that, by focusing excessively on export values, a rather misleading impression has been given of African performance. This paper brings to light some important, and often overlooked, characteristics to the determinants of African exports. Firstly, the impression created that African export performance in the 1990s was largely unsatisfactory is not correct. If by an unsatisfactory trade performance, we mean the apparent inability to move rapidly into new, high-value added, dynamic products, then African countries clearly failed. But, if we mean the capacity of African countries to expand their existing range of exports, then the argument is not so clear. Many African countries increased the volume of exports quite significantly during the reform period from the mid-1980s onwards. Secondly, for a multitude of reasons, African economies have not benefited as much as would be expected from the commodity price boom – in the period since 2002, the volume of exports have increased at a rate which has actually been slower than during the previous period of two decades of low and declining prices. Indeed, our econometric evidence would seem to suggest that export performance has been largely indifferent to price developments, and is more likely to be affected by policies which stimulate overall productivity and investment. Stabilising fragile states would also seem to be a priority if African countries are to become better integrated into the global economy. Finally, our analysis drives home the important point that it is not sufficient to export greater volumes – what matters is the ability to capture “rents”, in the Schumpeterian sense.11 Even if the. 11. See Kaplinsky (2005:Chapter 3) for a discussion of this issue.. 10.

(19) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. income terms of trade are positive, if technical progress is low in primary production, then growing export quantities may have high opportunity costs in resource terms. This means that the resources devoted to the production and export of these commodities could be used more effectively in other sectors (Kaplinsky, 2005:58). All this puts the importance of export diversification to the forefront. It must be noted that what diversification that has taken place in Sub-Saharan Africa took place before the period of reforms in the 1990s,12 and the subsequent commodity price boom has only consolidated trends towards greater concentration on primary exports. One positive consequence of the recent slump in commodity prices is that it will surely help focus minds once again on the need to diversify.. 12. See World Bank (2005: Table 3.6).. 11.

(20) Andrew Mold and Annalisa Prizzon. References ACKAH, C. and O. MORRISSEY (2005), “Trade Policy and Performance in Sub-Saharan Africa since the 1980s”, African Development Bank Economic Research Working Paper No. 78. ARELLANO, M. and O. BOVER, (1995), “Another Look at the Instrumental Variables Estimation of Error-Components Models,” Journal of Econometrics, No. 68, pp. 29-51. ARELLANO, M. and S. R. BOND, (1991), “Some Tests Of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations”, Review of Economic Studies, No. 58, pp. 277-297. BALASSA, B. (1989), “Economic Incentives and Agricultural Exports in Developing Countries”, in Nurul Islam (ed.), The Balance between Industry and Agriculture in Economic Development, pp. 181-196164-194, Macmillan, London. BLOCH, H. and D. SAPSFORD (2000), “Whither the Terms of Trade? An Elaboration of The Prebisch-Singer Hypothesis”, Cambridge Journal of Economics, No. 24, pp. 461–81. BOND, M.E. (1983), “Agricultural Responses to Prices in Sub-Saharan African Countries”, Staff Papers - International Monetary Fund, 30:4, pp. 703-726. BORATAV K. (2001), “Movement of Relative Agricultural Prices in Sub-Saharan Africa”, Cambridge Journal of Economics, No. 25, pp.395-416. CERRA, V. and S.C. SAXENA (2002), “An Empirical Analysis of China's Export Behavior”, IMF Working Papers, 02/2002, December. CHHIBBER, A. ( 1988), “Raising Agricultural Output: Price and Nonprice Factors”, Finance and Development, No. 25, pp. 44 47. CLARKE, G.R.G. (2005), “Beyond Tariffs and Quotas: Why Don’t African Manufactures Export More?”, World Bank Policy Research Working Paper, No. 3617, June. COE, D.T. and A. W. HOFFMAISTER (1998), “North-South Trade – Is Africa Unusual?” IMF Working Paper, WP98/94, June. COMMISSION FOR AFRICA (2005), Our Common Interest, Report of the Commission for Africa. Available at http://www.commissionforafrica.org/english/report/introduction.html. DFID (2005), Why We Need to Work More Effectively in Fragile States, London: DFID, January. EDWARDS, L. and P. ALVES (2006), “South Africa's Export Performance: Determinants of Export Supply”, South African Journal of Economics, Vol. 74, No. 3, pp. 473-500, September. FEENSTRA, R. (2004), Advanced International Trade, Princeton, N.J.: Princeton University Press. FOSU, A. (2003), “Political Instability and Export Performance in Sub-Saharan Africa”, Journal of Development Studies, Vol. 39, No. 4, pp. 68-83. GABRIELE, A. (1994), “Price Elasticity of Central American Agricultural Exports”, Cepal Review, 52, 105-114, April. GIBBON, P. and S. PONTE (2005), Trading Down: Africa, Value Chains and the Global Economy, Temple University Press, Philadelphia. GRILLI, E.R. And YANG M.C. (1988), “Commodity Prices, Manufactured Goods Prices and the Terms ofTrade of Developing Countries: What the Long Run Shows”, World Bank Economic Review, Vol. 2, No. 1. IMF (2008), World Economic Outlook, International Monetary Fund, April.. 12.

(21) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. KAPLINSKY, R. (2005), Globalization, Poverty and Inequality: Between a Rock and a Hard Place, Polity Press, Cambridge. KHANDELWA, P. (2004), “COMESA and SADC: Prospects and Challenges for Regional Trade Integration," IMF Working Papers, 04/227, International Monetary Fund. LINES, T. (2008), Making Poverty a History, Zed Book, London and New York. MAMINGI, N. (1996), “The Impact of Prices and Macroeconomic Policies on Agricultural Supply: a synthesis of available results”, Agricultural Economics, Vol. 16 , pp. 17-34. MBABAZI, J., C. MILNER and O. MORRISSEY (2003), “The Fragility Of Empirical Links Between Inequality, Trade Liberalisation, Growth And Poverty”, in R. van der Hoeven and A. Shorrocks (eds), Perspectives on Growth and Poverty, Tokyo and New York: United Nations University Press). MCKAY, A., O. MORRISSEY and C. VAILLANT (1997), “Trade Liberalisation and Agricultural Supply Response: Issues and Some Lessons”, European Journal of Development Research, Vol. 9, No. 2, pp. 129-147. MCKAY, A. and O. MORRISSEY and C. VAILLANT (1998), “Aggregate Export and Food Crop Supply Response in Tanzania”, University of Nottingham, CREDIT, Discussion Papers, 98/4. MOLD, A. (2004), “FDI and Poverty Reduction: A Critical Reappraisal of the Arguments”, Region et Developpement, No. 20, pp.61-83. MORRISSEY, O. (2005), “Imports and Implementation: Neglected Aspects of Trade in the Report of the Commission for Africa”, Journal of Development Studies, 41 (4), pp. 1133-1153. NAUDÉ, W. (2000), “The Determinants of South African Exports: An Econometric Analysis”, South African Journal of Economics, Vol. 68, No. 2, pp. 103-113, June. NG F. and A. YEATS (2002), “What Can Africa Expect From Is Its Traditional Exports?”, Africa Region Working Paper Series, No. 26, Washington D.C., the World Bank. NOORBAKHSH, F. and A. PALONI (1998), “Structural Adjustment Programmes and Export Supply Response”, Journal of International Development, Vol. 10, No. 4, 555-573. OCAMPO, J.A. and M.A. PARRA (2006), “The Commodity Terms Of Trade And Their Strategic Implications For Development”, in K.S. Jomo (ed.), Economic Globalization, Hegemony and the Changing World Economy during the Long Twentieth Century, pp. 164-194, Oxford University Press, New Delhi and Oxford. OGBU , O. M. and M. GBETIBOUO (1990), “Agricultural Supply Response in Sub-Saharan Africa: A Critical Review of the Literature”, African Development Review, Vol. 2, No. 2, pp. 83-99. ONAFOWORA, O. and O. OWOYE (1998), “Can Trade Liberalization Stimulate Economic Growth in Africa’, World Development, Vol. 26, No. 3, pp. 497-506. OZANNE, A. (1999), “Perverse Supply Response in Peasant Agriculture: A Review”, Oxford Development Studies, Vol. 27, No. 2. PREBISCH, P. (1950), The Economic Development of Latin America and Its Principal Problem. Santiago: UNECLA. RODRIK, D. (1999), The New Global Economy And Developing Countries: Making Openness Work, Washington, Overseas Development Council. SACHS, J.D. and A.M. WARNER (1997), “Sources of Slow Growth in African Economies”, Journal of African Economies, Vol. 6, No. 3, pp. 335-376. SANTOS-PAULINO, A. (2002a), “The Effect of Trade Liberalisation on Imports in Selected Developing Countries”, World Development, Vol. 30, No. 6 pp. 959-974.. 13.

(22) Andrew Mold and Annalisa Prizzon. SANTOS-PAULINO, A. (2002b), “Trade Liberalisation and Export Performance in Selected Developing Countries”, Journal of Development Studies, Vol. 39, No. 1, pp. 140-164. SINGER, H. (1950), “The Distribution of Gains between Investing and Borrowing Countries”, American Economic Review, Papers and Proceedings, Vol. 40, pp. 473–85. SÖDERBOM, M. and F. TEAL (2003), “Are Manufacturing Exports the Key to Economic Success in Africa?”, Journal of African Economies, Vol. 12, No. 1, pp. 1-29. SPRAOS, J. (1983), Inequalising Trade? A Study of Traditional North/South Specialisation in the Context of Terms of Trade Concepts, Oxford: Clarendon Press. UL HAQUE, I. (2004), “Commodities under Neoliberalism: The Case of Cocoa”, G-24 Discussion Paper Series, No. 25, January, Geneva, UNCTAD. UNCTAD (1999), African Development in a Comparative Perspective, Oxford, African World Press/James Currey. UNCTAD (2003), Developing Countries and World Trade – Performance and Prospects, edited by Yilmaz Akyuz, UNCTAD/TWN/Zed Books. UNCTAD (2008a), Economic Development in Africa - Export Performance Following Trade Liberalization: Some Patterns and Policy Perspectives, UNCTAD/GDS/AFRICA/2003/1 Geneva. UNCTAD (2008b), Trade and Development Report - Commodity Prices, Capital Flows and the Financing of Investment, Geneva and New York. UNCTAD (2008c), Handbook of Statistics, online version. WORLD BANK (2005), Economic Growth in the 1990s – Learning from a Decade of Reform, Washington, D.C. The World Bank. WORLD BANK (2008), Global Economic Prospects, Washington DC. WTO (2001), International Trade Statistics, Geneva: World Trade Organisation. WUYTS, M. (2005), Growth, Poverty Reduction and Terms of Trade: A Comment on Tanzania, mimeo, ISS, August. ZHENG Z. and Z. YUMIN (2002), “China’s terms of trade in manufactures, 1993-2000”, UNCTAD Discussion Paper, No. 161, June, Geneva.. 14.

(23) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. Appendix Table A.1. UNCTAD Export Volume Indexes of Export for African Countries 1990-2002 (100 = year 1990) Source: Calculated from UNCTAD Handbook of Statistics, 2004. Equatorial Guinea Mozambique Lesotho Uganda Sudan Central African Republic Mali Seychelles Burundi Guinea Bissau Zimbabwe Guinea United Republic of Tanzania Senegal Ethiopia Burkina Faso Congo Togo Tunisia Morocco Ghana Cape Verde Swaziland Kenya Mauritius Botswana South Africa Benin Namibia Algeria Zambia Cote d'Ivoire Mauritania Cameroon Comoros Malawi Dem Rep of the Congo Madagascar Gabon Nigeria Sao Tome and Principe Egypt Chad Rwanda Gambia Sierra Leone. 1990. 1994. 1998. 2002. 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0. 185.7 152.4 207.4 320.0 225.0 258.8 128.6 84.4 170.6 545.5 147.8 112.5 144.9 128.9 132.1 78.8 120.8 177.4 144.6 143.9 158.2 69.2 121.1 163.5 110.3 104.8 109.1 160.9 135.9 97.0 84.9 89.6 119.8 77.6 107.0 112.6 46.3 131.6 138.9 96.4 106.8 61.4 99.0 21.9 114.4 42.7. 1100.0 266.7 311.1 390.0 187.5 394.1 214.3 190.6 185.3 218.2 178.3 160.7 146.9 195.6 149.1 240.4 211.3 188.7 162.5 166.7 167.3 161.5 146.5 150.8 125.3 112.7 128.8 131.9 140.6 122.4 145.2 129.9 91.9 105.6 30.4 131.0 41.2 128.1 137.0 128.6 91.4 52.8 120.2 39.7 61.7 0.4. 3442.9 923.8 800.0 580.0 550.0 476.5 364.3 362.5 297.1 290.9 267.4 250.0 244.9 233.3 228.3 219.2 203.8 203.8 200.0 187.7 185.5 180.8 180.3 177.8 164.4 160.3 157.6 153.6 146.9 143.3 138.7 136.4 133.7 126.4 124.0 123.0 118.4 103.5 101.9 95.2 87.0 86.6 85.9 54.2 39.4 1.9. 15.

(24) Andrew Mold and Annalisa Prizzon. Table A.2. UNCTAD Export Volume Indices of Export for African Countries 2002-2006 (100 = year 2002). Source: Calculated from UNCTAD Handbook of Statistics, 2008. 2002. 2003. 2004. 2005. 2006. 2007. Chad. 100.0. 260.6. 747.1. 766.7. 694.9. 704.8. Sierra Leone. 100.0. 171.2. 225.4. 248.4. 308.7. 369.9. Uganda. 100.0. 117.2. 137.4. 141.9. 146.8. 216.0. Lesotho. 100.0. 132.9. 198.2. 181.5. 190.2. 206.7. Burkina Faso. 100.0. 104.4. 156.0. 157.6. 181.9. 188.3. Angola. 100.0. 99.0. 107.8. 137.9. 151.7. 177.2. Namibia. 100.0. 113.6. 151.8. 153.1. 160.8. 175.9. Mauritania. 100.0. 91.2. 113.7. 106.8. 194.0. 175.5. Equatorial Guinea. 100.0. 114.3. 145.1. 151.4. 154.5. 173.4. Tunisia. 100.0. 113.4. 130.9. 131.5. 138.5. 166.2. Zambia. 100.0. 94.8. 115.1. 108.0. 133.1. 164.9. Guinea-Bissau. 100.0. 116.0. 122.6. 138.3. 117.6. 163.2. Swaziland. 100.0. 155.1. 174.0. 165.6. 161.4. 162.2. Libyan Arab Jamahiriya. 100.0. 119.3. 115.9. 137.4. 155.5. 158.5. Mozambique. 100.0. 120.0. 148.8. 145.3. 155.8. 158.4. Ethiopia. 100.0. 94.7. 113.7. 128.9. 142.0. 155.1. Sudan. 100.0. 109.1. 130.5. 118.1. 118.0. 155.0. Egypt. 100.0. 117.8. 119.1. 130.3. 140.5. 153.5. Ghana. 100.0. 129.9. 145.4. 138.7. 160.3. 151.2. Malawi. 100.0. 127.9. 114.0. 110.6. 126.1. 149.4. Rwanda. 100.0. 89.8. 124.9. 106.9. 114.1. 127.7. Kenya. 100.0. 105.6. 105.4. 110.9. 110.0. 127.1. Congo, DRC. 100.0. 117.9. 145.1. 135.9. 121.1. 122.4. Cape Verde. 100.0. 118.5. 139.6. 132.4. 138.8. 119.7. Botswana. 100.0. 107.9. 125.3. 151.0. 137.8. 117.9. Nigeria. 100.0. 111.5. 125.3. 127.9. 117.3. 117.7. Morocco. 100.0. 95.8. 95.9. 98.6. 107.9. 117.3. Mauritius. 100.0. 121.7. 117.3. 127.8. 141.6. 116.6. South Africa. 100.0. 96.2. 101.4. 106.5. 103.1. 112.7. United Republic of Tanzania. 100.0. 111.6. 121.1. 126.6. 108.5. 112.3. Gabon. 100.0. 102.3. 104.2. 113.2. 105.9. 112.0. Congo. 100.0. 101.2. 101.4. 102.5. 119.4. 110.9. Côte d'Ivoire. 100.0. 105.5. 121.4. 114.1. 112.0. 105.3. Senegal. 100.0. 110.5. 120.9. 109.4. 98.4. 101.9. Madagascar. 100.0. 92.0. 106.2. 85.6. 92.9. 101.8. Benin. 100.0. 97.7. 102.6. 108.1. 97.6. 98.8. Seychelles. 100.0. 113.5. 104.2. 100.8. 107.3. 97.8. Central African Republic. 100.0. 85.5. 79.2. 78.5. 93.7. 95.7. Algeria. 100.0. 96.8. 105.5. 106.7. 98.2. 93.1. Burundi. 100.0. 106.4. 120.3. 116.2. 96.6. 91.0. 16.

(25) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. Table A.3. Country composition of Regional Areas Eastern Africa. Southern Africa. Central America. South Eastern Asia. Burundi Comoros Djibouti. Botswana Lesotho Namibia. Belize Costa Rica El Salvador. Eritrea Ethiopia Kenya Madagascar Malawi Mauritius Mayotte Mozambique Rwanda Seychelles Somalia Uganda Tanzania Zambia. South Africa Swaziland Western Africa Benin Burkina Faso Cape Verde Côte d'Ivoire Gambia Ghana Guinea Guinea-Bissau Liberia Mali Mauritania. Guatemala Honduras Mexico Nicaragua Panama South America Argentina Bolivia Brazil Chile Colombia Ecuador Falkland Islands (Malvinas) Guyana. Zimbabwe Middle Africa Angola Cameroon Central African Republic Chad Congo Congo, DRC Equatorial Guinea Gabon Sao Tome and Principe Northern Africa Algeria Egypt Libyan Arab Jamahiriya Morocco Sudan Tunisia. Niger Nigeria Saint Helena Senegal Sierra Leone Togo Caribbean Anguilla Antigua and Barbuda Aruba. Paraguay Peru Suriname Uruguay Venezuela (Bolivarian Republic of) Eastern Asia China Korea, Dem. Rep. Mongolia Republic of Korea. Brunei Darussalam Cambodia Indonesia Lao People's Democratic Republic Malaysia Myanmar Philippines Singapore Thailand Timor-Leste Viet Nam Western Asia Bahrain Iraq Jordan Kuwait Lebanon Occupied Palestinian territory Oman Qatar Saudi Arabia. Bahamas Barbados British Virgin Islands Cayman Islands. Southern Asia Afghanistan Bangladesh Bhutan. Cuba India including Sikkim Dominica Iran (Islamic Republic of) Dominican Republic Maldives Grenada Nepal Haiti Pakistan Jamaica Sri Lanka Montserrat Netherlands Antilles Saint Kitts and Nevis Saint Lucia Saint Vincent and the Grenadines Trinidad and Tobago. 17. Syrian Arab Republic Turkey United Arab Emirates Yemen.

(26) Andrew Mold and Annalisa Prizzon. Table A.4. Samples UNCTAD Database Angola Benin Botswana Burkina Faso Burundi Cameroon Central African Republic Chad Congo. Côte d'Ivoire Congo DRC Equatorial Guinea Ethiopia Gabon Gambia Ghana Guinea Guinea-Bissau. Angola Benin Burundi Cameroon Central African Republic Chad Comoros. Congo Congo DRC Ethiopia Gambia Guinea Guinea-Bissau Kenya. Kenya Lesotho Madagascar Malawi Mauritius Mozambique Niger Nigeria Rwanda. Senegal Seychelles Sierra Leone South Africa Sudan Togo Uganda Tanzania Zambia. Fragile countries Liberia Malawi Mali Mauritania Niger Nigeria Rwanda. Sao Tome and Principe Sierra Leone Somalia Sudan Togo Uganda Zimbabwe. Table A.5. Variables Variable. Description. Source. Volume_UNCTAD. Export Volume Index (100=yr2000). UNCTAD Handbook of Statistics. Price_UNCTAD. Export Unit Price Index (100=yr2000). UNCTAD Handbook of Statistics. OECD Import. Imports in USD High Income OECD Countries. World Development Indicators. REER. Real Effective Exchange Rate (100=yr2000). International Finance Statistics. GDPCAP. GDP per capita (LCU). World Development Indicators. FDI. Foreign Direct Investment (Inward) as a percentage of GDP. UNCTAD Handbook of Statistics. Annex A.6. Some Methodological Clarifications What economic meaning can be attached to the country-wide volume and unit price data indices reported in Section 2? The volume index for exports is calculated according to the Laspeyres formula as base-period-weighted averages and the unit price index for exports represents the familiar Paasche Index as current-period-weighted averages. T . The price and are the prices of commodity and are the quantities of ) are determined by ratio of the values of the individual commodities to the total commodity value of the group measured in current prices . The volume index (VOI) of a commodity group is a measure of the change in its value, which may be solely attributed to variations in the quantities of the individual commodities. The volume index is equal to:. 18.

(27) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. where is the price of commodity n at time 0, T is the number of commodities, and are the quantities of commodity n at time t and 0, respectively. The unit value indices13 are estimates of the unit values of total exports or imports from groups of countries in any given period, relative to the unit values of those exports or imports in a pre-determined based year. Similarly, the volume indices are estimates of the volume of total exports, by the same group of countries, in any period, relative to the volume of those exports in the base period. A.7. Fixed-effects Estimations. Dependent variable. Volume Index Unit Price Index. Exports Volume Index. Exports Volume Index. 0.90***. 0.76***. (56.27). (28.09). 0.06. -0.12. (1.11) OECD Imports. (-1.51) 0.13*** (3.36). REER. 0.01 (0.50). GDP per capita (+1). 0.41***. FDI Stock. 0.06***. (5.94) (3.54) Fragility interaction factor Time interaction factor. Number of observation. -0.04. -0.01. (-0.63). (-0.50). 0.01*. 0.00. (2.14). (-0.27). 1121. 759. Adjusted R-squared. 0.903172. 0.920647. S.E. of regression. 0.300192. 0.298887. 218.643. 210.3856. F-statistic. Prob(F-statistic) 0.0000 0.0000 Notes: t-statistics are reported under parenthesis. The levels of significance are as follows: *** 99 per cent level, **95 per cent level, *90 per cent level.. 13. See Methods used in compiling the United Nations price Indexed for External Trade Volume II. 19.

(28) Andrew Mold and Annalisa Prizzon. Table A. 8. Literature review on the price elasticity of export and supply functions Countries. Period. Scope and Methodology. Main Findings. Bond (1983). 6 SSA countries: Ivory Coast, Kenya, Malawi, Ghana, Sudan and Tanzania. 1954-56 1964-66 1975-77. Scope: Analysis of responsiveness of individual crop output (acreage under cultivation) to prices (real producer price) Methodology: Econometric Technique: OLS. - Both aggregate crop production and individual crops supply responses are positive - For individual crops price elasticity tends to be larger in the long rather than in the short-run - For aggregate crop there appear to be some evidence that farmers do respond to aggregate real producer prices. - No evidence for target- income hypothesis. Balassa (1988). 53 developing countries. 1965-82 (divided in two sub-periods) 1965-73 1974-82 (period of high versus period of low growth). Scope: Analysis of price incentives to agricultural and total exports Methodology: Econometric Technique: OLS. Total and agricultural exports are responsive to price incentives in SSA.. Scope: Critical review of the empirical literature on the behaviour of agricultural supply in SSA. - Binding constraints are not uniform across countries. - Cross-country estimation of supply functions suffers from the problem of establishing the direction of causality. The assumption underlying supply functions is that prices influence output or productivity, when, in fact, it is possible to argue that high agricultural productivity, which is associated with high per capita incomes, may lead to higher price support for agriculture. - The negative price elasticity is often associated with target income hypothesis and backward bending supply curve. While this is theoretically plausible from the work-leisure trade-off point of view, in the case of many SSA countries it can only be argued for the specific case of migrant seasonal workers or legislation concerning sale of labor services and can hardly be generalised. - The ability and willingness of a producer to react to favourable price changes or improved technology will depend on her/his ability to export as well as on land availability and farm-labour wage structure. - Factors that influence productivity such as technology, fertiliser use, institutional support, access to credit, health services, infrastructure, are more likely to influence output supply responsiveness to price.. Ogbu and Gbetibouo (1990). 20.

(29) Fragile States, Commodity Booms and Export Performance: An Analysis of the Sub-Saharan African Case. 1960-1990. Scope: Estimation of price elasticity of Central America agricultural exports Methodology: Simplified linear regression models with output prices as the only relevant explanatory variable for four crops bananas coffee sugar and cotton. No significant statistical relation is found between production and price in the cases of bananas and sugar, while cotton production appears to react to a certain extent only to year-to-year price changes.. Tanzania. 1964-1990. Scope: Analysis of the supply response of agricultural output in Tanzania Methodology: ECM. - Agricultural supply response is high. Potential for agricultural sector response to liberalisation of agricultural prices and marketing may be quite significant. - Long-run elasticity of food crop output to relative prices is almost unity. Both food and aggregate short-run response was estimated at about 0.35.. South Africa. 1974-1988. Scope: Analysis of the determinants of South African Exports Methodology: ECM. None of the determinants identified in the literature was found to be significant in explaining South Africa’s export volume.. China. Mid1980s-2001. Scope: Analysis of the determinants of China’s export volume Methodology: FE, 2SLS, Dynamic OLS adjusted for serial correlation and cross-sectional heterogeneity. Mixed results. However, regressions using aggregate export price and quantity indices clear indicate that export suppliers have increasingly behaved according to predictions of economic theory based upon market economy. In particular, the price elasticity of supply has become positive and increased over time, paralleling reforms in 1988 and 1991.. South Africa. 1982-99 1970-99 1980-99. Scope: Analysis of determinants of South Africa Manufacturing Export Performance Methodology: Dynamic Fixed effect model GMM difference and system. Exporters are responsive to polices and economic environments that raise the probability of export supply. Scope for policy makers to enhance substantially South Africa’s manufacturing export performance.. Gabriele (1994). Costa Rica, Salvador, Guatemala, Honduras. McKay, Morrissey, Vaillant (1998). Naudé (2000). Cerra Saxena (2002). and. Edwards and Alves (2006). El. 21.

(30) Andrew Mold and Annalisa Prizzon. Authors contacts: Andrew Mold OECD Development Centre, Paris 2, rue André Pascal F-75775 Paris Cedex 16 Email: Andrew.MOLD@oecd.org. Annalisa Prizzon OECD Development Centre, Paris 2, rue André Pascal F-75775 Paris Cedex 16 Email: Annalisa.PRIZZON@oecd.org. 22.

(31)